In [1]:
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
0. Problem biznesowy¶
Celem projektu jest klasteryzacja danych zebranych z czujników znajdjących się w smartfonach (akcelerometr, zyroskop, ...).
Projekt przewidywał stworzenie modelu przewidującego aktywności: chodzenie, wchodzenie po schodach, schodzenie schodami, siedzenie, stanie, leżenie.
Dataset został zebrany z 30 osób (później subjects) wykonujących wymienione aktywności.
1. wczytanie i podział danych¶
In [2]:
data = pd.read_csv('../data/data.csv')
In [3]:
from sklearn.model_selection import train_test_split
our_data, validator_data = train_test_split(data, test_size=0.2, random_state=42)
2. wstepna analiza¶
In [4]:
data.head()
Out[4]:
| tBodyAcc-mean()-X | tBodyAcc-mean()-Y | tBodyAcc-mean()-Z | tBodyAcc-std()-X | tBodyAcc-std()-Y | tBodyAcc-std()-Z | tBodyAcc-mad()-X | tBodyAcc-mad()-Y | tBodyAcc-mad()-Z | tBodyAcc-max()-X | ... | fBodyBodyGyroJerkMag-skewness() | fBodyBodyGyroJerkMag-kurtosis() | angle(tBodyAccMean,gravity) | angle(tBodyAccJerkMean),gravityMean) | angle(tBodyGyroMean,gravityMean) | angle(tBodyGyroJerkMean,gravityMean) | angle(X,gravityMean) | angle(Y,gravityMean) | angle(Z,gravityMean) | subject | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0.288585 | -0.020294 | -0.132905 | -0.995279 | -0.983111 | -0.913526 | -0.995112 | -0.983185 | -0.923527 | -0.934724 | ... | -0.298676 | -0.710304 | -0.112754 | 0.030400 | -0.464761 | -0.018446 | -0.841247 | 0.179941 | -0.058627 | 1 |
| 1 | 0.278419 | -0.016411 | -0.123520 | -0.998245 | -0.975300 | -0.960322 | -0.998807 | -0.974914 | -0.957686 | -0.943068 | ... | -0.595051 | -0.861499 | 0.053477 | -0.007435 | -0.732626 | 0.703511 | -0.844788 | 0.180289 | -0.054317 | 1 |
| 2 | 0.279653 | -0.019467 | -0.113462 | -0.995380 | -0.967187 | -0.978944 | -0.996520 | -0.963668 | -0.977469 | -0.938692 | ... | -0.390748 | -0.760104 | -0.118559 | 0.177899 | 0.100699 | 0.808529 | -0.848933 | 0.180637 | -0.049118 | 1 |
| 3 | 0.279174 | -0.026201 | -0.123283 | -0.996091 | -0.983403 | -0.990675 | -0.997099 | -0.982750 | -0.989302 | -0.938692 | ... | -0.117290 | -0.482845 | -0.036788 | -0.012892 | 0.640011 | -0.485366 | -0.848649 | 0.181935 | -0.047663 | 1 |
| 4 | 0.276629 | -0.016570 | -0.115362 | -0.998139 | -0.980817 | -0.990482 | -0.998321 | -0.979672 | -0.990441 | -0.942469 | ... | -0.351471 | -0.699205 | 0.123320 | 0.122542 | 0.693578 | -0.615971 | -0.847865 | 0.185151 | -0.043892 | 1 |
5 rows × 562 columns
In [7]:
firstrow = data.iloc[0]
for i in range(len(firstrow)):
print(firstrow.index[i], firstrow.iloc[i])
tBodyAcc-mean()-X 0.28858451 tBodyAcc-mean()-Y -0.020294171 tBodyAcc-mean()-Z -0.13290514 tBodyAcc-std()-X -0.9952786 tBodyAcc-std()-Y -0.98311061 tBodyAcc-std()-Z -0.91352645 tBodyAcc-mad()-X -0.99511208 tBodyAcc-mad()-Y -0.98318457 tBodyAcc-mad()-Z -0.92352702 tBodyAcc-max()-X -0.93472378 tBodyAcc-max()-Y -0.56737807 tBodyAcc-max()-Z -0.74441253 tBodyAcc-min()-X 0.85294738 tBodyAcc-min()-Y 0.68584458 tBodyAcc-min()-Z 0.81426278 tBodyAcc-sma() -0.96552279 tBodyAcc-energy()-X -0.99994465 tBodyAcc-energy()-Y -0.99986303 tBodyAcc-energy()-Z -0.99461218 tBodyAcc-iqr()-X -0.99423081 tBodyAcc-iqr()-Y -0.98761392 tBodyAcc-iqr()-Z -0.94321999 tBodyAcc-entropy()-X -0.40774707 tBodyAcc-entropy()-Y -0.67933751 tBodyAcc-entropy()-Z -0.60212187 tBodyAcc-arCoeff()-X,1 0.92929351 tBodyAcc-arCoeff()-X,2 -0.85301114 tBodyAcc-arCoeff()-X,3 0.35990976 tBodyAcc-arCoeff()-X,4 -0.058526382 tBodyAcc-arCoeff()-Y,1 0.25689154 tBodyAcc-arCoeff()-Y,2 -0.22484763 tBodyAcc-arCoeff()-Y,3 0.26410572 tBodyAcc-arCoeff()-Y,4 -0.09524563 tBodyAcc-arCoeff()-Z,1 0.27885143 tBodyAcc-arCoeff()-Z,2 -0.46508457 tBodyAcc-arCoeff()-Z,3 0.49193596 tBodyAcc-arCoeff()-Z,4 -0.19088356 tBodyAcc-correlation()-X,Y 0.37631389 tBodyAcc-correlation()-X,Z 0.43512919 tBodyAcc-correlation()-Y,Z 0.66079033 tGravityAcc-mean()-X 0.96339614 tGravityAcc-mean()-Y -0.14083968 tGravityAcc-mean()-Z 0.11537494 tGravityAcc-std()-X -0.98524969 tGravityAcc-std()-Y -0.98170843 tGravityAcc-std()-Z -0.87762497 tGravityAcc-mad()-X -0.98500137 tGravityAcc-mad()-Y -0.98441622 tGravityAcc-mad()-Z -0.89467735 tGravityAcc-max()-X 0.89205451 tGravityAcc-max()-Y -0.16126549 tGravityAcc-max()-Z 0.12465977 tGravityAcc-min()-X 0.97743631 tGravityAcc-min()-Y -0.12321341 tGravityAcc-min()-Z 0.056482734 tGravityAcc-sma() -0.37542596 tGravityAcc-energy()-X 0.89946864 tGravityAcc-energy()-Y -0.97090521 tGravityAcc-energy()-Z -0.97551037 tGravityAcc-iqr()-X -0.98432539 tGravityAcc-iqr()-Y -0.98884915 tGravityAcc-iqr()-Z -0.91774264 tGravityAcc-entropy()-X -1.0 tGravityAcc-entropy()-Y -1.0 tGravityAcc-entropy()-Z 0.11380614 tGravityAcc-arCoeff()-X,1 -0.590425 tGravityAcc-arCoeff()-X,2 0.5911463 tGravityAcc-arCoeff()-X,3 -0.59177346 tGravityAcc-arCoeff()-X,4 0.59246928 tGravityAcc-arCoeff()-Y,1 -0.74544878 tGravityAcc-arCoeff()-Y,2 0.72086167 tGravityAcc-arCoeff()-Y,3 -0.71237239 tGravityAcc-arCoeff()-Y,4 0.71130003 tGravityAcc-arCoeff()-Z,1 -0.99511159 tGravityAcc-arCoeff()-Z,2 0.99567491 tGravityAcc-arCoeff()-Z,3 -0.99566759 tGravityAcc-arCoeff()-Z,4 0.99165268 tGravityAcc-correlation()-X,Y 0.57022164 tGravityAcc-correlation()-X,Z 0.43902735 tGravityAcc-correlation()-Y,Z 0.98691312 tBodyAccJerk-mean()-X 0.077996345 tBodyAccJerk-mean()-Y 0.0050008031 tBodyAccJerk-mean()-Z -0.067830808 tBodyAccJerk-std()-X -0.99351906 tBodyAccJerk-std()-Y -0.98835999 tBodyAccJerk-std()-Z -0.99357497 tBodyAccJerk-mad()-X -0.99448763 tBodyAccJerk-mad()-Y -0.98620664 tBodyAccJerk-mad()-Z -0.99281835 tBodyAccJerk-max()-X -0.9851801 tBodyAccJerk-max()-Y -0.99199423 tBodyAccJerk-max()-Z -0.99311887 tBodyAccJerk-min()-X 0.98983471 tBodyAccJerk-min()-Y 0.99195686 tBodyAccJerk-min()-Z 0.9905192 tBodyAccJerk-sma() -0.99352201 tBodyAccJerk-energy()-X -0.99993487 tBodyAccJerk-energy()-Y -0.99982045 tBodyAccJerk-energy()-Z -0.99987846 tBodyAccJerk-iqr()-X -0.99436404 tBodyAccJerk-iqr()-Y -0.98602487 tBodyAccJerk-iqr()-Z -0.98923361 tBodyAccJerk-entropy()-X -0.81994925 tBodyAccJerk-entropy()-Y -0.79304645 tBodyAccJerk-entropy()-Z -0.88885295 tBodyAccJerk-arCoeff()-X,1 1.0 tBodyAccJerk-arCoeff()-X,2 -0.22074703 tBodyAccJerk-arCoeff()-X,3 0.63683075 tBodyAccJerk-arCoeff()-X,4 0.38764356 tBodyAccJerk-arCoeff()-Y,1 0.24140146 tBodyAccJerk-arCoeff()-Y,2 -0.052252848 tBodyAccJerk-arCoeff()-Y,3 0.2641772 tBodyAccJerk-arCoeff()-Y,4 0.37343945 tBodyAccJerk-arCoeff()-Z,1 0.34177752 tBodyAccJerk-arCoeff()-Z,2 -0.56979119 tBodyAccJerk-arCoeff()-Z,3 0.26539882 tBodyAccJerk-arCoeff()-Z,4 -0.47787489 tBodyAccJerk-correlation()-X,Y -0.3853005 tBodyAccJerk-correlation()-X,Z 0.033643943 tBodyAccJerk-correlation()-Y,Z -0.12651082 tBodyGyro-mean()-X -0.0061008489 tBodyGyro-mean()-Y -0.031364791 tBodyGyro-mean()-Z 0.1077254 tBodyGyro-std()-X -0.98531027 tBodyGyro-std()-Y -0.97662344 tBodyGyro-std()-Z -0.99220528 tBodyGyro-mad()-X -0.98458626 tBodyGyro-mad()-Y -0.97635262 tBodyGyro-mad()-Z -0.99236164 tBodyGyro-max()-X -0.86704374 tBodyGyro-max()-Y -0.93378602 tBodyGyro-max()-Z -0.74756618 tBodyGyro-min()-X 0.84730796 tBodyGyro-min()-Y 0.91489534 tBodyGyro-min()-Z 0.83084054 tBodyGyro-sma() -0.96718428 tBodyGyro-energy()-X -0.99957831 tBodyGyro-energy()-Y -0.99935432 tBodyGyro-energy()-Z -0.99976339 tBodyGyro-iqr()-X -0.98343808 tBodyGyro-iqr()-Y -0.97861401 tBodyGyro-iqr()-Z -0.99296558 tBodyGyro-entropy()-X 0.082631682 tBodyGyro-entropy()-Y 0.20226765 tBodyGyro-entropy()-Z -0.16875669 tBodyGyro-arCoeff()-X,1 0.096323236 tBodyGyro-arCoeff()-X,2 -0.27498511 tBodyGyro-arCoeff()-X,3 0.49864419 tBodyGyro-arCoeff()-X,4 -0.22031685 tBodyGyro-arCoeff()-Y,1 1.0 tBodyGyro-arCoeff()-Y,2 -0.97297139 tBodyGyro-arCoeff()-Y,3 0.31665451 tBodyGyro-arCoeff()-Y,4 0.37572641 tBodyGyro-arCoeff()-Z,1 0.72339919 tBodyGyro-arCoeff()-Z,2 -0.77111201 tBodyGyro-arCoeff()-Z,3 0.69021323 tBodyGyro-arCoeff()-Z,4 -0.33183104 tBodyGyro-correlation()-X,Y 0.70958377 tBodyGyro-correlation()-X,Z 0.13487336 tBodyGyro-correlation()-Y,Z 0.30109948 tBodyGyroJerk-mean()-X -0.0991674 tBodyGyroJerk-mean()-Y -0.055517369 tBodyGyroJerk-mean()-Z -0.061985797 tBodyGyroJerk-std()-X -0.99211067 tBodyGyroJerk-std()-Y -0.99251927 tBodyGyroJerk-std()-Z -0.99205528 tBodyGyroJerk-mad()-X -0.99216475 tBodyGyroJerk-mad()-Y -0.99494156 tBodyGyroJerk-mad()-Z -0.99261905 tBodyGyroJerk-max()-X -0.99015585 tBodyGyroJerk-max()-Y -0.98674277 tBodyGyroJerk-max()-Z -0.99204155 tBodyGyroJerk-min()-X 0.99442876 tBodyGyroJerk-min()-Y 0.99175581 tBodyGyroJerk-min()-Z 0.98935195 tBodyGyroJerk-sma() -0.99445335 tBodyGyroJerk-energy()-X -0.99993755 tBodyGyroJerk-energy()-Y -0.9999535 tBodyGyroJerk-energy()-Z -0.99992294 tBodyGyroJerk-iqr()-X -0.99229974 tBodyGyroJerk-iqr()-Y -0.99693892 tBodyGyroJerk-iqr()-Z -0.99224298 tBodyGyroJerk-entropy()-X -0.58985096 tBodyGyroJerk-entropy()-Y -0.68845905 tBodyGyroJerk-entropy()-Z -0.57210686 tBodyGyroJerk-arCoeff()-X,1 0.29237634 tBodyGyroJerk-arCoeff()-X,2 -0.36199802 tBodyGyroJerk-arCoeff()-X,3 0.40554269 tBodyGyroJerk-arCoeff()-X,4 -0.039006951 tBodyGyroJerk-arCoeff()-Y,1 0.98928381 tBodyGyroJerk-arCoeff()-Y,2 -0.41456048 tBodyGyroJerk-arCoeff()-Y,3 0.39160251 tBodyGyroJerk-arCoeff()-Y,4 0.28225087 tBodyGyroJerk-arCoeff()-Z,1 0.92726984 tBodyGyroJerk-arCoeff()-Z,2 -0.57237001 tBodyGyroJerk-arCoeff()-Z,3 0.6916192 tBodyGyroJerk-arCoeff()-Z,4 0.46828982 tBodyGyroJerk-correlation()-X,Y -0.13107697 tBodyGyroJerk-correlation()-X,Z -0.087159695 tBodyGyroJerk-correlation()-Y,Z 0.33624748 tBodyAccMag-mean() -0.95943388 tBodyAccMag-std() -0.9505515 tBodyAccMag-mad() -0.95799295 tBodyAccMag-max() -0.94630524 tBodyAccMag-min() -0.99255572 tBodyAccMag-sma() -0.95943388 tBodyAccMag-energy() -0.99849285 tBodyAccMag-iqr() -0.9576374 tBodyAccMag-entropy() -0.23258164 tBodyAccMag-arCoeff()1 -0.17317874 tBodyAccMag-arCoeff()2 -0.02289666 tBodyAccMag-arCoeff()3 0.094831568 tBodyAccMag-arCoeff()4 0.19181715 tGravityAccMag-mean() -0.95943388 tGravityAccMag-std() -0.9505515 tGravityAccMag-mad() -0.95799295 tGravityAccMag-max() -0.94630524 tGravityAccMag-min() -0.99255572 tGravityAccMag-sma() -0.95943388 tGravityAccMag-energy() -0.99849285 tGravityAccMag-iqr() -0.9576374 tGravityAccMag-entropy() -0.23258164 tGravityAccMag-arCoeff()1 -0.17317874 tGravityAccMag-arCoeff()2 -0.02289666 tGravityAccMag-arCoeff()3 0.094831568 tGravityAccMag-arCoeff()4 0.19181715 tBodyAccJerkMag-mean() -0.99330586 tBodyAccJerkMag-std() -0.99433641 tBodyAccJerkMag-mad() -0.99450037 tBodyAccJerkMag-max() -0.99278399 tBodyAccJerkMag-min() -0.99120847 tBodyAccJerkMag-sma() -0.99330586 tBodyAccJerkMag-energy() -0.99989188 tBodyAccJerkMag-iqr() -0.9929337 tBodyAccJerkMag-entropy() -0.86341476 tBodyAccJerkMag-arCoeff()1 0.28308522 tBodyAccJerkMag-arCoeff()2 -0.23730869 tBodyAccJerkMag-arCoeff()3 -0.10543219 tBodyAccJerkMag-arCoeff()4 -0.038212313 tBodyGyroMag-mean() -0.96895908 tBodyGyroMag-std() -0.96433518 tBodyGyroMag-mad() -0.95724477 tBodyGyroMag-max() -0.97505986 tBodyGyroMag-min() -0.99155366 tBodyGyroMag-sma() -0.96895908 tBodyGyroMag-energy() -0.99928646 tBodyGyroMag-iqr() -0.94976582 tBodyGyroMag-entropy() 0.072579035 tBodyGyroMag-arCoeff()1 0.57251142 tBodyGyroMag-arCoeff()2 -0.73860219 tBodyGyroMag-arCoeff()3 0.21257776 tBodyGyroMag-arCoeff()4 0.43340495 tBodyGyroJerkMag-mean() -0.99424782 tBodyGyroJerkMag-std() -0.99136761 tBodyGyroJerkMag-mad() -0.99314298 tBodyGyroJerkMag-max() -0.98893563 tBodyGyroJerkMag-min() -0.99348603 tBodyGyroJerkMag-sma() -0.99424782 tBodyGyroJerkMag-energy() -0.99994898 tBodyGyroJerkMag-iqr() -0.99454718 tBodyGyroJerkMag-entropy() -0.61976763 tBodyGyroJerkMag-arCoeff()1 0.29284049 tBodyGyroJerkMag-arCoeff()2 -0.1768892 tBodyGyroJerkMag-arCoeff()3 -0.14577921 tBodyGyroJerkMag-arCoeff()4 -0.12407233 fBodyAcc-mean()-X -0.99478319 fBodyAcc-mean()-Y -0.9829841 fBodyAcc-mean()-Z -0.93926865 fBodyAcc-std()-X -0.99542175 fBodyAcc-std()-Y -0.98313297 fBodyAcc-std()-Z -0.90616498 fBodyAcc-mad()-X -0.99688864 fBodyAcc-mad()-Y -0.98451927 fBodyAcc-mad()-Z -0.932082 fBodyAcc-max()-X -0.99375634 fBodyAcc-max()-Y -0.98316285 fBodyAcc-max()-Z -0.88505422 fBodyAcc-min()-X -0.99396185 fBodyAcc-min()-Y -0.99344611 fBodyAcc-min()-Z -0.92342772 fBodyAcc-sma() -0.97473271 fBodyAcc-energy()-X -0.99996838 fBodyAcc-energy()-Y -0.99968911 fBodyAcc-energy()-Z -0.99489148 fBodyAcc-iqr()-X -0.99592602 fBodyAcc-iqr()-Y -0.98970889 fBodyAcc-iqr()-Z -0.98799115 fBodyAcc-entropy()-X -0.94635692 fBodyAcc-entropy()-Y -0.90474776 fBodyAcc-entropy()-Z -0.59130248 fBodyAcc-maxInds-X -1.0 fBodyAcc-maxInds-Y -1.0 fBodyAcc-maxInds-Z -1.0 fBodyAcc-meanFreq()-X 0.2524829 fBodyAcc-meanFreq()-Y 0.13183575 fBodyAcc-meanFreq()-Z -0.052050251 fBodyAcc-skewness()-X 0.14205056 fBodyAcc-kurtosis()-X -0.1506825 fBodyAcc-skewness()-Y -0.22054694 fBodyAcc-kurtosis()-Y -0.55873853 fBodyAcc-skewness()-Z 0.24676868 fBodyAcc-kurtosis()-Z -0.0074155206 fBodyAcc-bandsEnergy()-1,8 -0.99996279 fBodyAcc-bandsEnergy()-9,16 -0.9999865 fBodyAcc-bandsEnergy()-17,24 -0.99997907 fBodyAcc-bandsEnergy()-25,32 -0.99996244 fBodyAcc-bandsEnergy()-33,40 -0.99993222 fBodyAcc-bandsEnergy()-41,48 -0.99972512 fBodyAcc-bandsEnergy()-49,56 -0.99967039 fBodyAcc-bandsEnergy()-57,64 -0.99998582 fBodyAcc-bandsEnergy()-1,16 -0.99996867 fBodyAcc-bandsEnergy()-17,32 -0.99997686 fBodyAcc-bandsEnergy()-33,48 -0.99986966 fBodyAcc-bandsEnergy()-49,64 -0.99977613 fBodyAcc-bandsEnergy()-1,24 -0.99997115 fBodyAcc-bandsEnergy()-25,48 -0.99991925 fBodyAcc-bandsEnergy()-1,8.1 -0.9996568 fBodyAcc-bandsEnergy()-9,16.1 -0.99986046 fBodyAcc-bandsEnergy()-17,24.1 -0.99986695 fBodyAcc-bandsEnergy()-25,32.1 -0.99986301 fBodyAcc-bandsEnergy()-33,40.1 -0.99973783 fBodyAcc-bandsEnergy()-41,48.1 -0.9997322 fBodyAcc-bandsEnergy()-49,56.1 -0.99949261 fBodyAcc-bandsEnergy()-57,64.1 -0.99981364 fBodyAcc-bandsEnergy()-1,16.1 -0.99968182 fBodyAcc-bandsEnergy()-17,32.1 -0.9998394 fBodyAcc-bandsEnergy()-33,48.1 -0.99973823 fBodyAcc-bandsEnergy()-49,64.1 -0.99961197 fBodyAcc-bandsEnergy()-1,24.1 -0.99968721 fBodyAcc-bandsEnergy()-25,48.1 -0.99983863 fBodyAcc-bandsEnergy()-1,8.2 -0.99359234 fBodyAcc-bandsEnergy()-9,16.2 -0.99947584 fBodyAcc-bandsEnergy()-17,24.2 -0.99966204 fBodyAcc-bandsEnergy()-25,32.2 -0.9996423 fBodyAcc-bandsEnergy()-33,40.2 -0.99929341 fBodyAcc-bandsEnergy()-41,48.2 -0.99789222 fBodyAcc-bandsEnergy()-49,56.2 -0.99593249 fBodyAcc-bandsEnergy()-57,64.2 -0.99514642 fBodyAcc-bandsEnergy()-1,16.2 -0.9947399 fBodyAcc-bandsEnergy()-17,32.2 -0.99968826 fBodyAcc-bandsEnergy()-33,48.2 -0.99892456 fBodyAcc-bandsEnergy()-49,64.2 -0.99567134 fBodyAcc-bandsEnergy()-1,24.2 -0.99487731 fBodyAcc-bandsEnergy()-25,48.2 -0.99945439 fBodyAccJerk-mean()-X -0.99233245 fBodyAccJerk-mean()-Y -0.98716991 fBodyAccJerk-mean()-Z -0.98969609 fBodyAccJerk-std()-X -0.99582068 fBodyAccJerk-std()-Y -0.99093631 fBodyAccJerk-std()-Z -0.99705167 fBodyAccJerk-mad()-X -0.99380547 fBodyAccJerk-mad()-Y -0.99051869 fBodyAccJerk-mad()-Z -0.99699279 fBodyAccJerk-max()-X -0.99673689 fBodyAccJerk-max()-Y -0.99197516 fBodyAccJerk-max()-Z -0.99324167 fBodyAccJerk-min()-X -0.99834907 fBodyAccJerk-min()-Y -0.99110842 fBodyAccJerk-min()-Z -0.95988537 fBodyAccJerk-sma() -0.99051499 fBodyAccJerk-energy()-X -0.99993475 fBodyAccJerk-energy()-Y -0.99982048 fBodyAccJerk-energy()-Z -0.99988449 fBodyAccJerk-iqr()-X -0.99302626 fBodyAccJerk-iqr()-Y -0.99137339 fBodyAccJerk-iqr()-Z -0.99623962 fBodyAccJerk-entropy()-X -1.0 fBodyAccJerk-entropy()-Y -1.0 fBodyAccJerk-entropy()-Z -1.0 fBodyAccJerk-maxInds-X 1.0 fBodyAccJerk-maxInds-Y -0.24 fBodyAccJerk-maxInds-Z -1.0 fBodyAccJerk-meanFreq()-X 0.87038451 fBodyAccJerk-meanFreq()-Y 0.210697 fBodyAccJerk-meanFreq()-Z 0.26370789 fBodyAccJerk-skewness()-X -0.70368577 fBodyAccJerk-kurtosis()-X -0.90374251 fBodyAccJerk-skewness()-Y -0.58257362 fBodyAccJerk-kurtosis()-Y -0.93631005 fBodyAccJerk-skewness()-Z -0.50734474 fBodyAccJerk-kurtosis()-Z -0.80553591 fBodyAccJerk-bandsEnergy()-1,8 -0.99998649 fBodyAccJerk-bandsEnergy()-9,16 -0.9999796 fBodyAccJerk-bandsEnergy()-17,24 -0.99997478 fBodyAccJerk-bandsEnergy()-25,32 -0.99995513 fBodyAccJerk-bandsEnergy()-33,40 -0.99991861 fBodyAccJerk-bandsEnergy()-41,48 -0.99964011 fBodyAccJerk-bandsEnergy()-49,56 -0.9994833 fBodyAccJerk-bandsEnergy()-57,64 -0.99996087 fBodyAccJerk-bandsEnergy()-1,16 -0.99998227 fBodyAccJerk-bandsEnergy()-17,32 -0.99997072 fBodyAccJerk-bandsEnergy()-33,48 -0.99981098 fBodyAccJerk-bandsEnergy()-49,64 -0.99948472 fBodyAccJerk-bandsEnergy()-1,24 -0.99998083 fBodyAccJerk-bandsEnergy()-25,48 -0.99985189 fBodyAccJerk-bandsEnergy()-1,8.1 -0.99993261 fBodyAccJerk-bandsEnergy()-9,16.1 -0.99989993 fBodyAccJerk-bandsEnergy()-17,24.1 -0.99982444 fBodyAccJerk-bandsEnergy()-25,32.1 -0.99985982 fBodyAccJerk-bandsEnergy()-33,40.1 -0.99972751 fBodyAccJerk-bandsEnergy()-41,48.1 -0.99972876 fBodyAccJerk-bandsEnergy()-49,56.1 -0.99956707 fBodyAccJerk-bandsEnergy()-57,64.1 -0.99976524 fBodyAccJerk-bandsEnergy()-1,16.1 -0.99990021 fBodyAccJerk-bandsEnergy()-17,32.1 -0.9998149 fBodyAccJerk-bandsEnergy()-33,48.1 -0.9997098 fBodyAccJerk-bandsEnergy()-49,64.1 -0.99959608 fBodyAccJerk-bandsEnergy()-1,24.1 -0.99985216 fBodyAccJerk-bandsEnergy()-25,48.1 -0.9998221 fBodyAccJerk-bandsEnergy()-1,8.2 -0.99939988 fBodyAccJerk-bandsEnergy()-9,16.2 -0.99976559 fBodyAccJerk-bandsEnergy()-17,24.2 -0.99995846 fBodyAccJerk-bandsEnergy()-25,32.2 -0.99994951 fBodyAccJerk-bandsEnergy()-33,40.2 -0.9998385 fBodyAccJerk-bandsEnergy()-41,48.2 -0.99981351 fBodyAccJerk-bandsEnergy()-49,56.2 -0.99878054 fBodyAccJerk-bandsEnergy()-57,64.2 -0.99857783 fBodyAccJerk-bandsEnergy()-1,16.2 -0.99961968 fBodyAccJerk-bandsEnergy()-17,32.2 -0.99998359 fBodyAccJerk-bandsEnergy()-33,48.2 -0.99982812 fBodyAccJerk-bandsEnergy()-49,64.2 -0.99868068 fBodyAccJerk-bandsEnergy()-1,24.2 -0.99984416 fBodyAccJerk-bandsEnergy()-25,48.2 -0.99992792 fBodyGyro-mean()-X -0.98657442 fBodyGyro-mean()-Y -0.98176153 fBodyGyro-mean()-Z -0.98951478 fBodyGyro-std()-X -0.98503264 fBodyGyro-std()-Y -0.97388607 fBodyGyro-std()-Z -0.99403493 fBodyGyro-mad()-X -0.98653085 fBodyGyro-mad()-Y -0.98361636 fBodyGyro-mad()-Z -0.99235201 fBodyGyro-max()-X -0.98049843 fBodyGyro-max()-Y -0.97227092 fBodyGyro-max()-Z -0.99494426 fBodyGyro-min()-X -0.99756862 fBodyGyro-min()-Y -0.9840851 fBodyGyro-min()-Z -0.99433541 fBodyGyro-sma() -0.98527621 fBodyGyro-energy()-X -0.99986371 fBodyGyro-energy()-Y -0.99966608 fBodyGyro-energy()-Z -0.99993462 fBodyGyro-iqr()-X -0.99034389 fBodyGyro-iqr()-Y -0.99483569 fBodyGyro-iqr()-Z -0.99441158 fBodyGyro-entropy()-X -0.71240225 fBodyGyro-entropy()-Y -0.64484236 fBodyGyro-entropy()-Z -0.83899298 fBodyGyro-maxInds-X -1.0 fBodyGyro-maxInds-Y -1.0 fBodyGyro-maxInds-Z -1.0 fBodyGyro-meanFreq()-X -0.25754888 fBodyGyro-meanFreq()-Y 0.097947109 fBodyGyro-meanFreq()-Z 0.54715105 fBodyGyro-skewness()-X 0.37731121 fBodyGyro-kurtosis()-X 0.13409154 fBodyGyro-skewness()-Y 0.27337197 fBodyGyro-kurtosis()-Y -0.091261831 fBodyGyro-skewness()-Z -0.4843465 fBodyGyro-kurtosis()-Z -0.7828507 fBodyGyro-bandsEnergy()-1,8 -0.99986502 fBodyGyro-bandsEnergy()-9,16 -0.99993178 fBodyGyro-bandsEnergy()-17,24 -0.99997295 fBodyGyro-bandsEnergy()-25,32 -0.99997018 fBodyGyro-bandsEnergy()-33,40 -0.99993012 fBodyGyro-bandsEnergy()-41,48 -0.99995862 fBodyGyro-bandsEnergy()-49,56 -0.99992899 fBodyGyro-bandsEnergy()-57,64 -0.99998465 fBodyGyro-bandsEnergy()-1,16 -0.99986326 fBodyGyro-bandsEnergy()-17,32 -0.99996815 fBodyGyro-bandsEnergy()-33,48 -0.9999361 fBodyGyro-bandsEnergy()-49,64 -0.99995363 fBodyGyro-bandsEnergy()-1,24 -0.99986442 fBodyGyro-bandsEnergy()-25,48 -0.99996098 fBodyGyro-bandsEnergy()-1,8.1 -0.99945373 fBodyGyro-bandsEnergy()-9,16.1 -0.99997811 fBodyGyro-bandsEnergy()-17,24.1 -0.99999153 fBodyGyro-bandsEnergy()-25,32.1 -0.9999901 fBodyGyro-bandsEnergy()-33,40.1 -0.99996857 fBodyGyro-bandsEnergy()-41,48.1 -0.99980657 fBodyGyro-bandsEnergy()-49,56.1 -0.998346 fBodyGyro-bandsEnergy()-57,64.1 -0.99896122 fBodyGyro-bandsEnergy()-1,16.1 -0.99961874 fBodyGyro-bandsEnergy()-17,32.1 -0.99998934 fBodyGyro-bandsEnergy()-33,48.1 -0.9999354 fBodyGyro-bandsEnergy()-49,64.1 -0.99838752 fBodyGyro-bandsEnergy()-1,24.1 -0.99964264 fBodyGyro-bandsEnergy()-25,48.1 -0.99997266 fBodyGyro-bandsEnergy()-1,8.2 -0.99995535 fBodyGyro-bandsEnergy()-9,16.2 -0.9999763 fBodyGyro-bandsEnergy()-17,24.2 -0.99990583 fBodyGyro-bandsEnergy()-25,32.2 -0.9999855 fBodyGyro-bandsEnergy()-33,40.2 -0.99993717 fBodyGyro-bandsEnergy()-41,48.2 -0.99975115 fBodyGyro-bandsEnergy()-49,56.2 -0.99907227 fBodyGyro-bandsEnergy()-57,64.2 -0.99992754 fBodyGyro-bandsEnergy()-1,16.2 -0.99995158 fBodyGyro-bandsEnergy()-17,32.2 -0.99990585 fBodyGyro-bandsEnergy()-33,48.2 -0.99989269 fBodyGyro-bandsEnergy()-49,64.2 -0.99944433 fBodyGyro-bandsEnergy()-1,24.2 -0.99994099 fBodyGyro-bandsEnergy()-25,48.2 -0.99995861 fBodyAccMag-mean() -0.95215466 fBodyAccMag-std() -0.95613397 fBodyAccMag-mad() -0.94887014 fBodyAccMag-max() -0.97432057 fBodyAccMag-min() -0.92572179 fBodyAccMag-sma() -0.95215466 fBodyAccMag-energy() -0.9982852 fBodyAccMag-iqr() -0.9732732 fBodyAccMag-entropy() -0.64637645 fBodyAccMag-maxInds -0.79310345 fBodyAccMag-meanFreq() -0.08843612 fBodyAccMag-skewness() -0.43647104 fBodyAccMag-kurtosis() -0.79684048 fBodyBodyAccJerkMag-mean() -0.99372565 fBodyBodyAccJerkMag-std() -0.99375495 fBodyBodyAccJerkMag-mad() -0.9919757 fBodyBodyAccJerkMag-max() -0.99336472 fBodyBodyAccJerkMag-min() -0.98817543 fBodyBodyAccJerkMag-sma() -0.99372565 fBodyBodyAccJerkMag-energy() -0.99991844 fBodyBodyAccJerkMag-iqr() -0.99136366 fBodyBodyAccJerkMag-entropy() -1.0 fBodyBodyAccJerkMag-maxInds -0.93650794 fBodyBodyAccJerkMag-meanFreq() 0.34698853 fBodyBodyAccJerkMag-skewness() -0.51608015 fBodyBodyAccJerkMag-kurtosis() -0.80276003 fBodyBodyGyroMag-mean() -0.98013485 fBodyBodyGyroMag-std() -0.96130944 fBodyBodyGyroMag-mad() -0.97365344 fBodyBodyGyroMag-max() -0.95226383 fBodyBodyGyroMag-min() -0.98949813 fBodyBodyGyroMag-sma() -0.98013485 fBodyBodyGyroMag-energy() -0.99924035 fBodyBodyGyroMag-iqr() -0.99265553 fBodyBodyGyroMag-entropy() -0.70129141 fBodyBodyGyroMag-maxInds -1.0 fBodyBodyGyroMag-meanFreq() -0.1289889 fBodyBodyGyroMag-skewness() 0.58615643 fBodyBodyGyroMag-kurtosis() 0.37460462 fBodyBodyGyroJerkMag-mean() -0.99199044 fBodyBodyGyroJerkMag-std() -0.99069746 fBodyBodyGyroJerkMag-mad() -0.98994084 fBodyBodyGyroJerkMag-max() -0.99244784 fBodyBodyGyroJerkMag-min() -0.99104773 fBodyBodyGyroJerkMag-sma() -0.99199044 fBodyBodyGyroJerkMag-energy() -0.99993676 fBodyBodyGyroJerkMag-iqr() -0.99045792 fBodyBodyGyroJerkMag-entropy() -0.8713058 fBodyBodyGyroJerkMag-maxInds -1.0 fBodyBodyGyroJerkMag-meanFreq() -0.074323027 fBodyBodyGyroJerkMag-skewness() -0.29867637 fBodyBodyGyroJerkMag-kurtosis() -0.71030407 angle(tBodyAccMean,gravity) -0.11275434 angle(tBodyAccJerkMean),gravityMean) 0.030400372 angle(tBodyGyroMean,gravityMean) -0.46476139 angle(tBodyGyroJerkMean,gravityMean) -0.018445884 angle(X,gravityMean) -0.84124676 angle(Y,gravityMean) 0.17994061 angle(Z,gravityMean) -0.058626924 subject 1.0
mega duzo kolumn
kolumna subject to numer ochotnika - raczej nielegalne informacje, trzeba wywalić
Jak zostały zebrane dane i co one oznaczają¶
Do zebrania danych zostały wykorzystane czujniki w telefonie (przyśpieszeniomierz, żyroskop), które zapisywały 3-osiowe przyśpieszenie liniowe(tAcc-XYZ) oraz 3-osiową prędkość kątową (tGyro-XYZ). Prefix 't' w metrykach oznacza czas.
Sygnał o przyśpieszeniu został rozdzielony na przyśpieszenie ciała i przyśpieszenie grawitacyjne (tBodyAcc-XYZ, tGravityAcc-XYZ).
Dodatkowo z przyśpieszenia i prędkości kątowej wyznaczono zryw (tBodyAccJerk-XYZ, tBodyGyroJerk-XYZ)
Obliczono wartość poszczególnych sygnałów przy pomocy normy Euklidesowej:
- tBodyAccMag
- tGravityAccMag
- tBodyAccJerkMag
- tBodyGyroMag
- tBodyGyroJerkMag
Prefix 'f' oznacza, że jest domeną frekwencyjną po zastosowaniu Szybkiej Transformacji Fouriera, czyli przekształcenia ciągu próbej sygnału $(a_0, a_1,..., a_{N-1})$, $a_i \in \R$ w ciąg harmoniczny $(A_0, A_1,...,A_{N-1})$, $A_i \in \R$, zgodne ze wzorem: $$ A_k = \sum^{N-1}_{n=0} a_n w_N ^{-kn}, 0\le k \le N-1, w_N = e^{i\frac{2\pi}{N}} $$ , gdzie i - liczba urojona, k - nr harmonicznej, n - nr próbki sygnału, $a_n$ - wartość próbki sygnału, N - liczba próbek.
Wszystkie dostępne sygnały:
- tBodyAcc
- tGravityAccMag
- tBodyGyroJerkMag
- tBodyAccJerk
- tBodyGyro
- tBodyAccJerkMag
- tBodyGyroJerk
- tBodyAccMag
- tGravityAcc
- tBodyGyroMag
- fBodyAcc
- fBodyBodyGyroMag
- fBodyAccJerk
- fBodyBodyAccJerkMag
- fBodyAccMag
- fBodyGyro
- fBodyBodyGyroJerkMag
In [53]:
data = data.drop('subject', axis=1)
our_data = our_data.drop('subject', axis=1)
validator_data = validator_data.drop('subject', axis=1)
In [54]:
our_data.to_csv('../data/our_data.csv', index=False)
validator_data.to_csv('../data/validator_data.csv', index=False)
In [55]:
data.info()
<class 'pandas.core.frame.DataFrame'> RangeIndex: 10299 entries, 0 to 10298 Columns: 561 entries, tBodyAcc-mean()-X to angle(Z,gravityMean) dtypes: float64(561) memory usage: 44.1 MB
In [56]:
data.shape
Out[56]:
(10299, 561)
In [57]:
data.describe()
Out[57]:
| tBodyAcc-mean()-X | tBodyAcc-mean()-Y | tBodyAcc-mean()-Z | tBodyAcc-std()-X | tBodyAcc-std()-Y | tBodyAcc-std()-Z | tBodyAcc-mad()-X | tBodyAcc-mad()-Y | tBodyAcc-mad()-Z | tBodyAcc-max()-X | ... | fBodyBodyGyroJerkMag-meanFreq() | fBodyBodyGyroJerkMag-skewness() | fBodyBodyGyroJerkMag-kurtosis() | angle(tBodyAccMean,gravity) | angle(tBodyAccJerkMean),gravityMean) | angle(tBodyGyroMean,gravityMean) | angle(tBodyGyroJerkMean,gravityMean) | angle(X,gravityMean) | angle(Y,gravityMean) | angle(Z,gravityMean) | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| count | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | ... | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 | 10299.000000 |
| mean | 0.274347 | -0.017743 | -0.108925 | -0.607784 | -0.510191 | -0.613064 | -0.633593 | -0.525697 | -0.614989 | -0.466732 | ... | 0.126708 | -0.298592 | -0.617700 | 0.007705 | 0.002648 | 0.017683 | -0.009219 | -0.496522 | 0.063255 | -0.054284 |
| std | 0.067628 | 0.037128 | 0.053033 | 0.438694 | 0.500240 | 0.403657 | 0.413333 | 0.484201 | 0.399034 | 0.538707 | ... | 0.245443 | 0.320199 | 0.308796 | 0.336591 | 0.447364 | 0.616188 | 0.484770 | 0.511158 | 0.305468 | 0.268898 |
| min | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | ... | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 | -1.000000 |
| 25% | 0.262625 | -0.024902 | -0.121019 | -0.992360 | -0.976990 | -0.979137 | -0.993293 | -0.977017 | -0.979064 | -0.935788 | ... | -0.019481 | -0.536174 | -0.841847 | -0.124694 | -0.287031 | -0.493108 | -0.389041 | -0.817288 | 0.002151 | -0.131880 |
| 50% | 0.277174 | -0.017162 | -0.108596 | -0.943030 | -0.835032 | -0.850773 | -0.948244 | -0.843670 | -0.845068 | -0.874825 | ... | 0.136245 | -0.335160 | -0.703402 | 0.008146 | 0.007668 | 0.017192 | -0.007186 | -0.715631 | 0.182028 | -0.003882 |
| 75% | 0.288354 | -0.010625 | -0.097589 | -0.250293 | -0.057336 | -0.278737 | -0.302033 | -0.087405 | -0.288149 | -0.014641 | ... | 0.288960 | -0.113167 | -0.487981 | 0.149005 | 0.291490 | 0.536137 | 0.365996 | -0.521503 | 0.250790 | 0.102970 |
| max | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | ... | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 |
8 rows × 561 columns
In [58]:
data.isna().sum().sum()
Out[58]:
0
In [59]:
data.duplicated().sum()
Out[59]:
0
In [60]:
data = our_data
nie ma duplikatow, nie ma wartości NA
In [61]:
#columns with variances
for column in data.columns:
print(column + ' variance: ' + str(data[column].var()))
tBodyAcc-mean()-X variance: 0.0044769421018401164 tBodyAcc-mean()-Y variance: 0.0013073837539268314 tBodyAcc-mean()-Z variance: 0.0027694192760906685 tBodyAcc-std()-X variance: 0.1925145874633604 tBodyAcc-std()-Y variance: 0.2500234056668236 tBodyAcc-std()-Z variance: 0.16317495619877742 tBodyAcc-mad()-X variance: 0.17104109761769018 tBodyAcc-mad()-Y variance: 0.23392930786117078 tBodyAcc-mad()-Z variance: 0.15925895593970477 tBodyAcc-max()-X variance: 0.28980073061210443 tBodyAcc-max()-Y variance: 0.0783974216221772 tBodyAcc-max()-Z variance: 0.08009845958439489 tBodyAcc-min()-X variance: 0.12626623345624427 tBodyAcc-min()-Y variance: 0.11516030415949183 tBodyAcc-min()-Z variance: 0.0851457185762241 tBodyAcc-sma() variance: 0.21282353332234252 tBodyAcc-energy()-X variance: 0.06130929887210224 tBodyAcc-energy()-Y variance: 0.015620590506499237 tBodyAcc-energy()-Z variance: 0.042629911965941326 tBodyAcc-iqr()-X variance: 0.12929907140304242 tBodyAcc-iqr()-Y variance: 0.13550404180670667 tBodyAcc-iqr()-Z variance: 0.13802162331687137 tBodyAcc-entropy()-X variance: 0.2140492037240372 tBodyAcc-entropy()-Y variance: 0.18732332852241745 tBodyAcc-entropy()-Z variance: 0.1342684321248004 tBodyAcc-arCoeff()-X,1 variance: 0.0948414383496001 tBodyAcc-arCoeff()-X,2 variance: 0.06079429393582888 tBodyAcc-arCoeff()-X,3 variance: 0.061074745199654765 tBodyAcc-arCoeff()-X,4 variance: 0.053390542150113955 tBodyAcc-arCoeff()-Y,1 variance: 0.06388026695475865 tBodyAcc-arCoeff()-Y,2 variance: 0.04507507775145629 tBodyAcc-arCoeff()-Y,3 variance: 0.0433245157407126 tBodyAcc-arCoeff()-Y,4 variance: 0.04872749089065405 tBodyAcc-arCoeff()-Z,1 variance: 0.07890871360908631 tBodyAcc-arCoeff()-Z,2 variance: 0.046315234156252244 tBodyAcc-arCoeff()-Z,3 variance: 0.05574323892116306 tBodyAcc-arCoeff()-Z,4 variance: 0.05217893517859387 tBodyAcc-correlation()-X,Y variance: 0.12868741850115376 tBodyAcc-correlation()-X,Z variance: 0.10613257911568273 tBodyAcc-correlation()-Y,Z variance: 0.1428173950084769 tGravityAcc-mean()-X variance: 0.26119283211115724 tGravityAcc-mean()-Y variance: 0.14354652444117244 tGravityAcc-mean()-Z variance: 0.11058150218461149 tGravityAcc-std()-X variance: 0.005585243951767919 tGravityAcc-std()-Y variance: 0.007087955712434068 tGravityAcc-std()-Z variance: 0.009984117451470528 tGravityAcc-mad()-X variance: 0.005374241928028323 tGravityAcc-mad()-Y variance: 0.006889667280456189 tGravityAcc-mad()-Z variance: 0.009769062416794544 tGravityAcc-max()-X variance: 0.2547809711493905 tGravityAcc-max()-Y variance: 0.13435920566499165 tGravityAcc-max()-Z variance: 0.10749808670944473 tGravityAcc-min()-X variance: 0.252576622207885 tGravityAcc-min()-Y variance: 0.14165065753858258 tGravityAcc-min()-Z variance: 0.11222901950221918 tGravityAcc-sma() variance: 0.15221271912207224 tGravityAcc-energy()-X variance: 0.48062549291042816 tGravityAcc-energy()-Y variance: 0.20633463735234914 tGravityAcc-energy()-Z variance: 0.17606999508840906 tGravityAcc-iqr()-X variance: 0.004823924624644773 tGravityAcc-iqr()-Y variance: 0.0062561661734055005 tGravityAcc-iqr()-Z variance: 0.008874297130694766 tGravityAcc-entropy()-X variance: 0.13060044204300483 tGravityAcc-entropy()-Y variance: 0.07881709359491207 tGravityAcc-entropy()-Z variance: 0.15494499076632529 tGravityAcc-arCoeff()-X,1 variance: 0.046616644723727695 tGravityAcc-arCoeff()-X,2 variance: 0.043864112824116196 tGravityAcc-arCoeff()-X,3 variance: 0.043114806063649994 tGravityAcc-arCoeff()-X,4 variance: 0.044294547152750136 tGravityAcc-arCoeff()-Y,1 variance: 0.08510984181073537 tGravityAcc-arCoeff()-Y,2 variance: 0.08839127806474463 tGravityAcc-arCoeff()-Y,3 variance: 0.08375534481092596 tGravityAcc-arCoeff()-Y,4 variance: 0.07883361156891049 tGravityAcc-arCoeff()-Z,1 variance: 0.07164088372801314 tGravityAcc-arCoeff()-Z,2 variance: 0.0669291046694802 tGravityAcc-arCoeff()-Z,3 variance: 0.06378614451312242 tGravityAcc-arCoeff()-Z,4 variance: 0.06198864602083171 tGravityAcc-correlation()-X,Y variance: 0.48611656625829164 tGravityAcc-correlation()-X,Z variance: 0.4962658502056412 tGravityAcc-correlation()-Y,Z variance: 0.49241811393933693 tBodyAccJerk-mean()-X variance: 0.03099463167602009 tBodyAccJerk-mean()-Y variance: 0.027195642907117294 tBodyAccJerk-mean()-Z variance: 0.024249179038239525 tBodyAccJerk-std()-X variance: 0.16586271638033623 tBodyAccJerk-std()-Y variance: 0.18728524287794135 tBodyAccJerk-std()-Z variance: 0.07682020849228101 tBodyAccJerk-mad()-X variance: 0.17012896033680533 tBodyAccJerk-mad()-Y variance: 0.19913841884393885 tBodyAccJerk-mad()-Z variance: 0.08004200516158354 tBodyAccJerk-max()-X variance: 0.12793846731804348 tBodyAccJerk-max()-Y variance: 0.08676480507159709 tBodyAccJerk-max()-Z variance: 0.0499912875678026 tBodyAccJerk-min()-X variance: 0.18877009088054233 tBodyAccJerk-min()-Y variance: 0.12803282275044528 tBodyAccJerk-min()-Z variance: 0.09391796938758086 tBodyAccJerk-sma() variance: 0.15372057621725846 tBodyAccJerk-energy()-X variance: 0.0457138945027856 tBodyAccJerk-energy()-Y variance: 0.05957802170614774 tBodyAccJerk-energy()-Z variance: 0.013629326774191835 tBodyAccJerk-iqr()-X variance: 0.18390578349998593 tBodyAccJerk-iqr()-Y variance: 0.1409118403820026 tBodyAccJerk-iqr()-Z variance: 0.07024445818496423 tBodyAccJerk-entropy()-X variance: 0.4250574027599599 tBodyAccJerk-entropy()-Y variance: 0.3986348977807441 tBodyAccJerk-entropy()-Z variance: 0.3693758412362016 tBodyAccJerk-arCoeff()-X,1 variance: 0.09201627955021724 tBodyAccJerk-arCoeff()-X,2 variance: 0.03688828804159097 tBodyAccJerk-arCoeff()-X,3 variance: 0.05841863245116189 tBodyAccJerk-arCoeff()-X,4 variance: 0.03857935466853554 tBodyAccJerk-arCoeff()-Y,1 variance: 0.07097740700165883 tBodyAccJerk-arCoeff()-Y,2 variance: 0.04704483168131824 tBodyAccJerk-arCoeff()-Y,3 variance: 0.056390837814098184 tBodyAccJerk-arCoeff()-Y,4 variance: 0.04123905739477185 tBodyAccJerk-arCoeff()-Z,1 variance: 0.07327766822967441 tBodyAccJerk-arCoeff()-Z,2 variance: 0.038650895989761944 tBodyAccJerk-arCoeff()-Z,3 variance: 0.04932771198321671 tBodyAccJerk-arCoeff()-Z,4 variance: 0.057422860147315256 tBodyAccJerk-correlation()-X,Y variance: 0.06732082198080272 tBodyAccJerk-correlation()-X,Z variance: 0.08583053219679816 tBodyAccJerk-correlation()-Y,Z variance: 0.0757253265572673 tBodyGyro-mean()-X variance: 0.03401948423535753 tBodyGyro-mean()-Y variance: 0.01792968572581448 tBodyGyro-mean()-Z variance: 0.01795841962651189 tBodyGyro-std()-X variance: 0.09107630944742055 tBodyGyro-std()-Y variance: 0.125861263259997 tBodyGyro-std()-Z variance: 0.13950567110311113 tBodyGyro-mad()-X variance: 0.08903099584342154 tBodyGyro-mad()-Y variance: 0.11722642426721408 tBodyGyro-mad()-Z variance: 0.13168508711009588 tBodyGyro-max()-X variance: 0.07706050712775221 tBodyGyro-max()-Y variance: 0.069807631350714 tBodyGyro-max()-Z variance: 0.09267567176064244 tBodyGyro-min()-X variance: 0.06376422996141676 tBodyGyro-min()-Y variance: 0.04405604125578248 tBodyGyro-min()-Z variance: 0.09491938398868191 tBodyGyro-sma() variance: 0.16168717836590835 tBodyGyro-energy()-X variance: 0.020808135990968247 tBodyGyro-energy()-Y variance: 0.03526885118499151 tBodyGyro-energy()-Z variance: 0.03650740730891167 tBodyGyro-iqr()-X variance: 0.0918052894489914 tBodyGyro-iqr()-Y variance: 0.1043406325563918 tBodyGyro-iqr()-Z variance: 0.10108679847957205 tBodyGyro-entropy()-X variance: 0.20731747806158213 tBodyGyro-entropy()-Y variance: 0.13696450966881693 tBodyGyro-entropy()-Z variance: 0.20672442174307706 tBodyGyro-arCoeff()-X,1 variance: 0.07592136562353444 tBodyGyro-arCoeff()-X,2 variance: 0.05135225460885689 tBodyGyro-arCoeff()-X,3 variance: 0.05259559757937087 tBodyGyro-arCoeff()-X,4 variance: 0.05860014260699311 tBodyGyro-arCoeff()-Y,1 variance: 0.04308135822783237 tBodyGyro-arCoeff()-Y,2 variance: 0.03641994810051895 tBodyGyro-arCoeff()-Y,3 variance: 0.05168538864069485 tBodyGyro-arCoeff()-Y,4 variance: 0.04677002823097223 tBodyGyro-arCoeff()-Z,1 variance: 0.09789678928597803 tBodyGyro-arCoeff()-Z,2 variance: 0.07988933444151035 tBodyGyro-arCoeff()-Z,3 variance: 0.0720508560467873 tBodyGyro-arCoeff()-Z,4 variance: 0.06684542471175238 tBodyGyro-correlation()-X,Y variance: 0.14479692971951272 tBodyGyro-correlation()-X,Z variance: 0.1481607260735439 tBodyGyro-correlation()-Y,Z variance: 0.17328354951440553 tBodyGyroJerk-mean()-X variance: 0.016621110926878954 tBodyGyroJerk-mean()-Y variance: 0.012932609901319824 tBodyGyroJerk-mean()-Z variance: 0.01639661687251594 tBodyGyroJerk-std()-X variance: 0.0930770473744336 tBodyGyroJerk-std()-Y variance: 0.07328262480478664 tBodyGyroJerk-std()-Z variance: 0.09138512190072078 tBodyGyroJerk-mad()-X variance: 0.09468324819797247 tBodyGyroJerk-mad()-Y variance: 0.06539268706068921 tBodyGyroJerk-mad()-Z variance: 0.0857992152838401 tBodyGyroJerk-max()-X variance: 0.08585521799590187 tBodyGyroJerk-max()-Y variance: 0.06344211474922835 tBodyGyroJerk-max()-Z variance: 0.09421256287884862 tBodyGyroJerk-min()-X variance: 0.07901016181827171 tBodyGyroJerk-min()-Y variance: 0.050676412316502537 tBodyGyroJerk-min()-Z variance: 0.062150133786041506 tBodyGyroJerk-sma() variance: 0.07211516548214168 tBodyGyroJerk-energy()-X variance: 0.018295011568934764 tBodyGyroJerk-energy()-Y variance: 0.017040429017863276 tBodyGyroJerk-energy()-Z variance: 0.01815356215178411 tBodyGyroJerk-iqr()-X variance: 0.09168856438444567 tBodyGyroJerk-iqr()-Y variance: 0.057684769513897616 tBodyGyroJerk-iqr()-Z variance: 0.0736965517625947 tBodyGyroJerk-entropy()-X variance: 0.3176628578059097 tBodyGyroJerk-entropy()-Y variance: 0.29635770107721965 tBodyGyroJerk-entropy()-Z variance: 0.3368254087879385 tBodyGyroJerk-arCoeff()-X,1 variance: 0.061868345053234346 tBodyGyroJerk-arCoeff()-X,2 variance: 0.034653556125095096 tBodyGyroJerk-arCoeff()-X,3 variance: 0.044275946396582076 tBodyGyroJerk-arCoeff()-X,4 variance: 0.04479865526389249 tBodyGyroJerk-arCoeff()-Y,1 variance: 0.0452957285943773 tBodyGyroJerk-arCoeff()-Y,2 variance: 0.027872675612099287 tBodyGyroJerk-arCoeff()-Y,3 variance: 0.038732745112459076 tBodyGyroJerk-arCoeff()-Y,4 variance: 0.06032404169887483 tBodyGyroJerk-arCoeff()-Z,1 variance: 0.11796076463468294 tBodyGyroJerk-arCoeff()-Z,2 variance: 0.05395404829324188 tBodyGyroJerk-arCoeff()-Z,3 variance: 0.06108062584572923 tBodyGyroJerk-arCoeff()-Z,4 variance: 0.06155361666695325 tBodyGyroJerk-correlation()-X,Y variance: 0.07631687791784077 tBodyGyroJerk-correlation()-X,Z variance: 0.07027767624536363 tBodyGyroJerk-correlation()-Y,Z variance: 0.06737083576597896 tBodyAccMag-mean() variance: 0.21821054021982741 tBodyAccMag-std() variance: 0.18409706053673333 tBodyAccMag-mad() variance: 0.1399736194502291 tBodyAccMag-max() variance: 0.2132807468001808 tBodyAccMag-min() variance: 0.03620316454253036 tBodyAccMag-sma() variance: 0.21821054021982741 tBodyAccMag-energy() variance: 0.07878119529689388 tBodyAccMag-iqr() variance: 0.09832731854155205 tBodyAccMag-entropy() variance: 0.4468045665430065 tBodyAccMag-arCoeff()1 variance: 0.08401775130928239 tBodyAccMag-arCoeff()2 variance: 0.05483315918939998 tBodyAccMag-arCoeff()3 variance: 0.0637065419808146 tBodyAccMag-arCoeff()4 variance: 0.0688129863209937 tGravityAccMag-mean() variance: 0.21821054021982741 tGravityAccMag-std() variance: 0.18409706053673333 tGravityAccMag-mad() variance: 0.1399736194502291 tGravityAccMag-max() variance: 0.2132807468001808 tGravityAccMag-min() variance: 0.03620316454253036 tGravityAccMag-sma() variance: 0.21821054021982741 tGravityAccMag-energy() variance: 0.07878119529689388 tGravityAccMag-iqr() variance: 0.09832731854155205 tGravityAccMag-entropy() variance: 0.4468045665430065 tGravityAccMag-arCoeff()1 variance: 0.08401775130928239 tGravityAccMag-arCoeff()2 variance: 0.05483315918939998 tGravityAccMag-arCoeff()3 variance: 0.0637065419808146 tGravityAccMag-arCoeff()4 variance: 0.0688129863209937 tBodyAccJerkMag-mean() variance: 0.1510826893974949 tBodyAccJerkMag-std() variance: 0.17286870285130101 tBodyAccJerkMag-mad() variance: 0.15712103611348635 tBodyAccJerkMag-max() variance: 0.1619681171686266 tBodyAccJerkMag-min() variance: 0.0650885145222443 tBodyAccJerkMag-sma() variance: 0.1510826893974949 tBodyAccJerkMag-energy() variance: 0.04215218986476785 tBodyAccJerkMag-iqr() variance: 0.11710930812181335 tBodyAccJerkMag-entropy() variance: 0.523554748660506 tBodyAccJerkMag-arCoeff()1 variance: 0.05911154612879264 tBodyAccJerkMag-arCoeff()2 variance: 0.05982163496941966 tBodyAccJerkMag-arCoeff()3 variance: 0.05786532442764301 tBodyAccJerkMag-arCoeff()4 variance: 0.06735846772629919 tBodyGyroMag-mean() variance: 0.1596720200912777 tBodyGyroMag-std() variance: 0.12182080683711213 tBodyGyroMag-mad() variance: 0.1448849093236813 tBodyGyroMag-max() variance: 0.10161664772248963 tBodyGyroMag-min() variance: 0.09769292705225238 tBodyGyroMag-sma() variance: 0.1596720200912777 tBodyGyroMag-energy() variance: 0.04888900730292003 tBodyGyroMag-iqr() variance: 0.1295348600033986 tBodyGyroMag-entropy() variance: 0.23246526369405152 tBodyGyroMag-arCoeff()1 variance: 0.08757477369718733 tBodyGyroMag-arCoeff()2 variance: 0.07090210745304783 tBodyGyroMag-arCoeff()3 variance: 0.06333372149805352 tBodyGyroMag-arCoeff()4 variance: 0.06272022110858624 tBodyGyroJerkMag-mean() variance: 0.07593704083188658 tBodyGyroJerkMag-std() variance: 0.07323890086652206 tBodyGyroJerkMag-mad() variance: 0.062495061328663504 tBodyGyroJerkMag-max() variance: 0.07115244574100077 tBodyGyroJerkMag-min() variance: 0.06156129858268948 tBodyGyroJerkMag-sma() variance: 0.07593704083188658 tBodyGyroJerkMag-energy() variance: 0.014498435295459765 tBodyGyroJerkMag-iqr() variance: 0.05339057651073691 tBodyGyroJerkMag-entropy() variance: 0.4786859948086381 tBodyGyroJerkMag-arCoeff()1 variance: 0.0570737254544036 tBodyGyroJerkMag-arCoeff()2 variance: 0.043057561592672204 tBodyGyroJerkMag-arCoeff()3 variance: 0.051603635343599685 tBodyGyroJerkMag-arCoeff()4 variance: 0.06264963013681654 fBodyAcc-mean()-X variance: 0.17620049108501215 fBodyAcc-mean()-Y variance: 0.23155664211097182 fBodyAcc-mean()-Z variance: 0.12836606043181653 fBodyAcc-std()-X variance: 0.20004254496403195 fBodyAcc-std()-Y variance: 0.23027176888969544 fBodyAcc-std()-Z variance: 0.159757365148621 fBodyAcc-mad()-X variance: 0.20867076373107318 fBodyAcc-mad()-Y variance: 0.2461145771701687 fBodyAcc-mad()-Z variance: 0.1516610920170035 fBodyAcc-max()-X variance: 0.16564466518248028 fBodyAcc-max()-Y variance: 0.12507511408126354 fBodyAcc-max()-Z variance: 0.1528670770366282 fBodyAcc-min()-X variance: 0.05103581883158596 fBodyAcc-min()-Y variance: 0.032412894324738026 fBodyAcc-min()-Z variance: 0.01678977083504041 fBodyAcc-sma() variance: 0.22177023922421213 fBodyAcc-energy()-X variance: 0.06126143004211884 fBodyAcc-energy()-Y variance: 0.09840695485178359 fBodyAcc-energy()-Z variance: 0.05130099867037763 fBodyAcc-iqr()-X variance: 0.1580760886029029 fBodyAcc-iqr()-Y variance: 0.15231597573274552 fBodyAcc-iqr()-Z variance: 0.08401167760880253 fBodyAcc-entropy()-X variance: 0.5210502075320648 fBodyAcc-entropy()-Y variance: 0.44179422771404436 fBodyAcc-entropy()-Z variance: 0.3718011918685037 fBodyAcc-maxInds-X variance: 0.0698670126235585 fBodyAcc-maxInds-Y variance: 0.05950292122980049 fBodyAcc-maxInds-Z variance: 0.058359215219027344 fBodyAcc-meanFreq()-X variance: 0.07024375838333395 fBodyAcc-meanFreq()-Y variance: 0.05757650895276057 fBodyAcc-meanFreq()-Z variance: 0.08072816365251509 fBodyAcc-skewness()-X variance: 0.1613047195828785 fBodyAcc-kurtosis()-X variance: 0.1952393941181204 fBodyAcc-skewness()-Y variance: 0.12720204087308673 fBodyAcc-kurtosis()-Y variance: 0.14891024412338724 fBodyAcc-skewness()-Z variance: 0.1604929403017108 fBodyAcc-kurtosis()-Z variance: 0.16898194833628355 fBodyAcc-bandsEnergy()-1,8 variance: 0.07238759870011621 fBodyAcc-bandsEnergy()-9,16 variance: 0.03080378133830322 fBodyAcc-bandsEnergy()-17,24 variance: 0.049938829899136596 fBodyAcc-bandsEnergy()-25,32 variance: 0.0327164438372514 fBodyAcc-bandsEnergy()-33,40 variance: 0.019809664351880207 fBodyAcc-bandsEnergy()-41,48 variance: 0.02230304474886189 fBodyAcc-bandsEnergy()-49,56 variance: 0.010628570655686087 fBodyAcc-bandsEnergy()-57,64 variance: 0.014105546086979231 fBodyAcc-bandsEnergy()-1,16 variance: 0.06635876633757386 fBodyAcc-bandsEnergy()-17,32 variance: 0.05593857031857847 fBodyAcc-bandsEnergy()-33,48 variance: 0.019348748260329368 fBodyAcc-bandsEnergy()-49,64 variance: 0.01098265273977148 fBodyAcc-bandsEnergy()-1,24 variance: 0.06273730671383304 fBodyAcc-bandsEnergy()-25,48 variance: 0.03547599612806119 fBodyAcc-bandsEnergy()-1,8.1 variance: 0.07481340630992561 fBodyAcc-bandsEnergy()-9,16.1 variance: 0.05739621269958564 fBodyAcc-bandsEnergy()-17,24.1 variance: 0.048402879865634026 fBodyAcc-bandsEnergy()-25,32.1 variance: 0.02715588234719452 fBodyAcc-bandsEnergy()-33,40.1 variance: 0.03033634693683529 fBodyAcc-bandsEnergy()-41,48.1 variance: 0.0350775634366622 fBodyAcc-bandsEnergy()-49,56.1 variance: 0.028304092888693906 fBodyAcc-bandsEnergy()-57,64.1 variance: 0.016780546638190855 fBodyAcc-bandsEnergy()-1,16.1 variance: 0.0950071730465999 fBodyAcc-bandsEnergy()-17,32.1 variance: 0.06319810702624862 fBodyAcc-bandsEnergy()-33,48.1 variance: 0.03651580950555846 fBodyAcc-bandsEnergy()-49,64.1 variance: 0.022524363599338973 fBodyAcc-bandsEnergy()-1,24.1 variance: 0.09675182878395555 fBodyAcc-bandsEnergy()-25,48.1 variance: 0.030368017558191154 fBodyAcc-bandsEnergy()-1,8.2 variance: 0.04675605921555085 fBodyAcc-bandsEnergy()-9,16.2 variance: 0.03070034859912075 fBodyAcc-bandsEnergy()-17,24.2 variance: 0.020542299591896913 fBodyAcc-bandsEnergy()-25,32.2 variance: 0.006182878541287872 fBodyAcc-bandsEnergy()-33,40.2 variance: 0.00560915266223008 fBodyAcc-bandsEnergy()-41,48.2 variance: 0.013541715825299254 fBodyAcc-bandsEnergy()-49,56.2 variance: 0.011581790203825508 fBodyAcc-bandsEnergy()-57,64.2 variance: 0.012650742698531309 fBodyAcc-bandsEnergy()-1,16.2 variance: 0.045218015512478865 fBodyAcc-bandsEnergy()-17,32.2 variance: 0.013189430430581068 fBodyAcc-bandsEnergy()-33,48.2 variance: 0.007683831698815254 fBodyAcc-bandsEnergy()-49,64.2 variance: 0.010758590645850101 fBodyAcc-bandsEnergy()-1,24.2 variance: 0.049643888598150644 fBodyAcc-bandsEnergy()-25,48.2 variance: 0.00619283227209873 fBodyAccJerk-mean()-X variance: 0.15070652197924725 fBodyAccJerk-mean()-Y variance: 0.16575641178018785 fBodyAccJerk-mean()-Z variance: 0.08808846556388308 fBodyAccJerk-std()-X variance: 0.1539147603940635 fBodyAccJerk-std()-Y variance: 0.18788415998897146 fBodyAccJerk-std()-Z variance: 0.06707721373877917 fBodyAccJerk-mad()-X variance: 0.20971935812533224 fBodyAccJerk-mad()-Y variance: 0.17465976890554846 fBodyAccJerk-mad()-Z variance: 0.07599133900603627 fBodyAccJerk-max()-X variance: 0.11444653339280422 fBodyAccJerk-max()-Y variance: 0.1330133911965325 fBodyAccJerk-max()-Z variance: 0.05939345066234401 fBodyAccJerk-min()-X variance: 0.03472798016560612 fBodyAccJerk-min()-Y variance: 0.046227192269697626 fBodyAccJerk-min()-Z variance: 0.032143264059482805 fBodyAccJerk-sma() variance: 0.17634307429820362 fBodyAccJerk-energy()-X variance: 0.045833443454325884 fBodyAccJerk-energy()-Y variance: 0.05953324883268387 fBodyAccJerk-energy()-Z variance: 0.013623591865885964 fBodyAccJerk-iqr()-X variance: 0.17542697482174513 fBodyAccJerk-iqr()-Y variance: 0.09768761956082063 fBodyAccJerk-iqr()-Z variance: 0.07062446493967557 fBodyAccJerk-entropy()-X variance: 0.5600704138790288 fBodyAccJerk-entropy()-Y variance: 0.5379347564460603 fBodyAccJerk-entropy()-Z variance: 0.4074890648988625 fBodyAccJerk-maxInds-X variance: 0.10435294515414725 fBodyAccJerk-maxInds-Y variance: 0.0701849696967341 fBodyAccJerk-maxInds-Z variance: 0.08590974988802902 fBodyAccJerk-meanFreq()-X variance: 0.08739592533398391 fBodyAccJerk-meanFreq()-Y variance: 0.07317232424028697 fBodyAccJerk-meanFreq()-Z variance: 0.07409022611740908 fBodyAccJerk-skewness()-X variance: 0.0668323160129934 fBodyAccJerk-kurtosis()-X variance: 0.04468564147062609 fBodyAccJerk-skewness()-Y variance: 0.03557914507541991 fBodyAccJerk-kurtosis()-Y variance: 0.02017944807478049 fBodyAccJerk-skewness()-Z variance: 0.04020468871774056 fBodyAccJerk-kurtosis()-Z variance: 0.022946699718231094 fBodyAccJerk-bandsEnergy()-1,8 variance: 0.04453097847763921 fBodyAccJerk-bandsEnergy()-9,16 variance: 0.029847697165176386 fBodyAccJerk-bandsEnergy()-17,24 variance: 0.040477624611083514 fBodyAccJerk-bandsEnergy()-25,32 variance: 0.031775988800925185 fBodyAccJerk-bandsEnergy()-33,40 variance: 0.017138559412520935 fBodyAccJerk-bandsEnergy()-41,48 variance: 0.026387746743091425 fBodyAccJerk-bandsEnergy()-49,56 variance: 0.010221083942420798 fBodyAccJerk-bandsEnergy()-57,64 variance: 0.002714807081025747 fBodyAccJerk-bandsEnergy()-1,16 variance: 0.03736419960926148 fBodyAccJerk-bandsEnergy()-17,32 variance: 0.05156040977746052 fBodyAccJerk-bandsEnergy()-33,48 variance: 0.02209469475939296 fBodyAccJerk-bandsEnergy()-49,64 variance: 0.010945234046959375 fBodyAccJerk-bandsEnergy()-1,24 variance: 0.0478666130924042 fBodyAccJerk-bandsEnergy()-25,48 variance: 0.04922959832269986 fBodyAccJerk-bandsEnergy()-1,8.1 variance: 0.055596688420752265 fBodyAccJerk-bandsEnergy()-9,16.1 variance: 0.043169098220662935 fBodyAccJerk-bandsEnergy()-17,24.1 variance: 0.06631824503418884 fBodyAccJerk-bandsEnergy()-25,32.1 variance: 0.024668069216722516 fBodyAccJerk-bandsEnergy()-33,40.1 variance: 0.02130546077584993 fBodyAccJerk-bandsEnergy()-41,48.1 variance: 0.03992551319742578 fBodyAccJerk-bandsEnergy()-49,56.1 variance: 0.01677312955398876 fBodyAccJerk-bandsEnergy()-57,64.1 variance: 0.008648954970739113 fBodyAccJerk-bandsEnergy()-1,16.1 variance: 0.05437786140775746 fBodyAccJerk-bandsEnergy()-17,32.1 variance: 0.06304735090615839 fBodyAccJerk-bandsEnergy()-33,48.1 variance: 0.03794310199287232 fBodyAccJerk-bandsEnergy()-49,64.1 variance: 0.014242531673827313 fBodyAccJerk-bandsEnergy()-1,24.1 variance: 0.07097448187585614 fBodyAccJerk-bandsEnergy()-25,48.1 variance: 0.02765759482644583 fBodyAccJerk-bandsEnergy()-1,8.2 variance: 0.027101885895900537 fBodyAccJerk-bandsEnergy()-9,16.2 variance: 0.03154979231033198 fBodyAccJerk-bandsEnergy()-17,24.2 variance: 0.01729144970949675 fBodyAccJerk-bandsEnergy()-25,32.2 variance: 0.005945491394425677 fBodyAccJerk-bandsEnergy()-33,40.2 variance: 0.004995985986168579 fBodyAccJerk-bandsEnergy()-41,48.2 variance: 0.012377866437112504 fBodyAccJerk-bandsEnergy()-49,56.2 variance: 0.01665979715089601 fBodyAccJerk-bandsEnergy()-57,64.2 variance: 0.009016311990507937 fBodyAccJerk-bandsEnergy()-1,16.2 variance: 0.039427941862873975 fBodyAccJerk-bandsEnergy()-17,32.2 variance: 0.010009085664863982 fBodyAccJerk-bandsEnergy()-33,48.2 variance: 0.0071265437493440095 fBodyAccJerk-bandsEnergy()-49,64.2 variance: 0.016477365995092034 fBodyAccJerk-bandsEnergy()-1,24.2 variance: 0.02799157512760521 fBodyAccJerk-bandsEnergy()-25,48.2 variance: 0.006005011449334394 fBodyGyro-mean()-X variance: 0.12425841988085169 fBodyGyro-mean()-Y variance: 0.1110442062267809 fBodyGyro-mean()-Z variance: 0.1458722357091164 fBodyGyro-std()-X variance: 0.08206923622471465 fBodyGyro-std()-Y variance: 0.1352346541271663 fBodyGyro-std()-Z variance: 0.11457627364762704 fBodyGyro-mad()-X variance: 0.11230251123240358 fBodyGyro-mad()-Y variance: 0.10489264471168488 fBodyGyro-mad()-Z variance: 0.1490763532763312 fBodyGyro-max()-X variance: 0.09285927639359372 fBodyGyro-max()-Y variance: 0.10533026682020001 fBodyGyro-max()-Z variance: 0.07604083660126311 fBodyGyro-min()-X variance: 0.012754629686748826 fBodyGyro-min()-Y variance: 0.02438893932017824 fBodyGyro-min()-Z variance: 0.019763815780703804 fBodyGyro-sma() variance: 0.1303211429464363 fBodyGyro-energy()-X variance: 0.017152848182698962 fBodyGyro-energy()-Y variance: 0.035628924884671696 fBodyGyro-energy()-Z variance: 0.03987962996582353 fBodyGyro-iqr()-X variance: 0.11378132487225272 fBodyGyro-iqr()-Y variance: 0.09097429241179016 fBodyGyro-iqr()-Z variance: 0.1236435730609421 fBodyGyro-entropy()-X variance: 0.369748691485735 fBodyGyro-entropy()-Y variance: 0.37137297651690737 fBodyGyro-entropy()-Z variance: 0.3625669175668779 fBodyGyro-maxInds-X variance: 0.03596993479702613 fBodyGyro-maxInds-Y variance: 0.08267228756467052 fBodyGyro-maxInds-Z variance: 0.054324438085144186 fBodyGyro-meanFreq()-X variance: 0.06482591398246966 fBodyGyro-meanFreq()-Y variance: 0.07460712542629397 fBodyGyro-meanFreq()-Z variance: 0.0704103881337236 fBodyGyro-skewness()-X variance: 0.10437364560641044 fBodyGyro-kurtosis()-X variance: 0.11588116724449628 fBodyGyro-skewness()-Y variance: 0.12045898335047683 fBodyGyro-kurtosis()-Y variance: 0.14251109840825638 fBodyGyro-skewness()-Z variance: 0.11029025929318306 fBodyGyro-kurtosis()-Z variance: 0.12105992554498364 fBodyGyro-bandsEnergy()-1,8 variance: 0.013963062026664218 fBodyGyro-bandsEnergy()-9,16 variance: 0.025364034345867544 fBodyGyro-bandsEnergy()-17,24 variance: 0.019586007957239297 fBodyGyro-bandsEnergy()-25,32 variance: 0.006882683692325504 fBodyGyro-bandsEnergy()-33,40 variance: 0.01184724710025043 fBodyGyro-bandsEnergy()-41,48 variance: 0.008642375051335825 fBodyGyro-bandsEnergy()-49,56 variance: 0.006714892424487017 fBodyGyro-bandsEnergy()-57,64 variance: 0.005317954900239434 fBodyGyro-bandsEnergy()-1,16 variance: 0.015982640309287095 fBodyGyro-bandsEnergy()-17,32 variance: 0.01962906407320969 fBodyGyro-bandsEnergy()-33,48 variance: 0.011978962709321772 fBodyGyro-bandsEnergy()-49,64 variance: 0.005826304477666091 fBodyGyro-bandsEnergy()-1,24 variance: 0.016711515867859034 fBodyGyro-bandsEnergy()-25,48 variance: 0.007863112754786376 fBodyGyro-bandsEnergy()-1,8.1 variance: 0.046545591422753996 fBodyGyro-bandsEnergy()-9,16.1 variance: 0.010149007851245817 fBodyGyro-bandsEnergy()-17,24.1 variance: 0.01261057045658394 fBodyGyro-bandsEnergy()-25,32.1 variance: 0.006896208147389761 fBodyGyro-bandsEnergy()-33,40.1 variance: 0.0034802287752631 fBodyGyro-bandsEnergy()-41,48.1 variance: 0.008868098891297877 fBodyGyro-bandsEnergy()-49,56.1 variance: 0.012267887474151123 fBodyGyro-bandsEnergy()-57,64.1 variance: 0.00580507368294781 fBodyGyro-bandsEnergy()-1,16.1 variance: 0.03360665018625851 fBodyGyro-bandsEnergy()-17,32.1 variance: 0.01527430221000299 fBodyGyro-bandsEnergy()-33,48.1 variance: 0.004216412026657845 fBodyGyro-bandsEnergy()-49,64.1 variance: 0.01094511102445687 fBodyGyro-bandsEnergy()-1,24.1 variance: 0.03999738713786341 fBodyGyro-bandsEnergy()-25,48.1 variance: 0.00666430455750655 fBodyGyro-bandsEnergy()-1,8.2 variance: 0.029193088250484136 fBodyGyro-bandsEnergy()-9,16.2 variance: 0.017927759057571362 fBodyGyro-bandsEnergy()-17,24.2 variance: 0.01888437609020458 fBodyGyro-bandsEnergy()-25,32.2 variance: 0.005648181206527622 fBodyGyro-bandsEnergy()-33,40.2 variance: 0.004348621931241796 fBodyGyro-bandsEnergy()-41,48.2 variance: 0.006176934335451651 fBodyGyro-bandsEnergy()-49,56.2 variance: 0.011047976352854429 fBodyGyro-bandsEnergy()-57,64.2 variance: 0.008329489549744438 fBodyGyro-bandsEnergy()-1,16.2 variance: 0.03542011369842754 fBodyGyro-bandsEnergy()-17,32.2 variance: 0.025295047038539444 fBodyGyro-bandsEnergy()-33,48.2 variance: 0.004563521776722529 fBodyGyro-bandsEnergy()-49,64.2 variance: 0.009202730974715238 fBodyGyro-bandsEnergy()-1,24.2 variance: 0.03848340888221707 fBodyGyro-bandsEnergy()-25,48.2 variance: 0.005004281723665741 fBodyAccMag-mean() variance: 0.19744241991363984 fBodyAccMag-std() variance: 0.12635038451167352 fBodyAccMag-mad() variance: 0.18415740883923254 fBodyAccMag-max() variance: 0.0667298625060265 fBodyAccMag-min() variance: 0.024877929116656414 fBodyAccMag-sma() variance: 0.19744241991363984 fBodyAccMag-energy() variance: 0.06082934541673548 fBodyAccMag-iqr() variance: 0.11894280797441864 fBodyAccMag-entropy() variance: 0.45897389838240277 fBodyAccMag-maxInds variance: 0.06850248458427 fBodyAccMag-meanFreq() variance: 0.06882951452210465 fBodyAccMag-skewness() variance: 0.10271468146537477 fBodyAccMag-kurtosis() variance: 0.10093035713954598 fBodyBodyAccJerkMag-mean() variance: 0.1794370276310095 fBodyBodyAccJerkMag-std() variance: 0.1637197641921945 fBodyBodyAccJerkMag-mad() variance: 0.1844863731768773 fBodyBodyAccJerkMag-max() variance: 0.13627663760994213 fBodyBodyAccJerkMag-min() variance: 0.07210586132914354 fBodyBodyAccJerkMag-sma() variance: 0.1794370276310095 fBodyBodyAccJerkMag-energy() variance: 0.051498013073888814 fBodyBodyAccJerkMag-iqr() variance: 0.13310524980230382 fBodyBodyAccJerkMag-entropy() variance: 0.44312360155605013 fBodyBodyAccJerkMag-maxInds variance: 0.034980593548474444 fBodyBodyAccJerkMag-meanFreq() variance: 0.06288259066473362 fBodyBodyAccJerkMag-skewness() variance: 0.13379221142709152 fBodyBodyAccJerkMag-kurtosis() variance: 0.1261789218892486 fBodyBodyGyroMag-mean() variance: 0.10432726722704622 fBodyBodyGyroMag-std() variance: 0.09648458368892042 fBodyBodyGyroMag-mad() variance: 0.10976948261690145 fBodyBodyGyroMag-max() variance: 0.07907077036280058 fBodyBodyGyroMag-min() variance: 0.026751208252962448 fBodyBodyGyroMag-sma() variance: 0.10432726722704622 fBodyBodyGyroMag-energy() variance: 0.03260213889669768 fBodyBodyGyroMag-iqr() variance: 0.09614727223887459 fBodyBodyGyroMag-entropy() variance: 0.3624935745221581 fBodyBodyGyroMag-maxInds variance: 0.02558548810345983 fBodyBodyGyroMag-meanFreq() variance: 0.07804496389090335 fBodyBodyGyroMag-skewness() variance: 0.10339994669655914 fBodyBodyGyroMag-kurtosis() variance: 0.10195292018655154 fBodyBodyGyroJerkMag-mean() variance: 0.07080176375343458 fBodyBodyGyroJerkMag-std() variance: 0.06709234324886033 fBodyBodyGyroJerkMag-mad() variance: 0.07800145906977704 fBodyBodyGyroJerkMag-max() variance: 0.0589548954432414 fBodyBodyGyroJerkMag-min() variance: 0.03634454108884741 fBodyBodyGyroJerkMag-sma() variance: 0.07080176375343458 fBodyBodyGyroJerkMag-energy() variance: 0.01626321832853924 fBodyBodyGyroJerkMag-iqr() variance: 0.0770973458674192 fBodyBodyGyroJerkMag-entropy() variance: 0.38826912122814766 fBodyBodyGyroJerkMag-maxInds variance: 0.01990033006089468 fBodyBodyGyroJerkMag-meanFreq() variance: 0.059722875548427326 fBodyBodyGyroJerkMag-skewness() variance: 0.10423887776141551 fBodyBodyGyroJerkMag-kurtosis() variance: 0.09736195823596573 angle(tBodyAccMean,gravity) variance: 0.11353362835650334 angle(tBodyAccJerkMean),gravityMean) variance: 0.19998342356266013 angle(tBodyGyroMean,gravityMean) variance: 0.38381257790020445 angle(tBodyGyroJerkMean,gravityMean) variance: 0.238903350439246 angle(X,gravityMean) variance: 0.2570861163389007 angle(Y,gravityMean) variance: 0.09344515621016526 angle(Z,gravityMean) variance: 0.07138322841986906
moze usunąć, te kolumny, ktore mają 'małą' wariancję?
In [62]:
for column in data.columns:
if data[column].var() < 0.005:
print(column + ' variance: ' + str(data[column].var()))
tBodyAcc-mean()-X variance: 0.0044769421018401164 tBodyAcc-mean()-Y variance: 0.0013073837539268314 tBodyAcc-mean()-Z variance: 0.0027694192760906685 tGravityAcc-iqr()-X variance: 0.004823924624644773 fBodyAccJerk-bandsEnergy()-57,64 variance: 0.002714807081025747 fBodyAccJerk-bandsEnergy()-33,40.2 variance: 0.004995985986168579 fBodyGyro-bandsEnergy()-33,40.1 variance: 0.0034802287752631 fBodyGyro-bandsEnergy()-33,48.1 variance: 0.004216412026657845 fBodyGyro-bandsEnergy()-33,40.2 variance: 0.004348621931241796 fBodyGyro-bandsEnergy()-33,48.2 variance: 0.004563521776722529
zadna z kolumn nie ma wariancji = 0, trudno stwierdzic czy warto usuwac
In [63]:
atributes = set()
static_methods = set()
for col in data.columns:
atributes.add(col.split("-")[0])
try:
static_methods.add(col.split("-")[1])
except IndexError:
continue
for atr in atributes:
print(atr + "\n")
tBodyAccMag tBodyAccJerk angle(tBodyAccJerkMean),gravityMean) tBodyAcc angle(Y,gravityMean) fBodyAcc tBodyGyroMag tBodyAccJerkMag fBodyGyro angle(tBodyGyroMean,gravityMean) fBodyAccMag tGravityAcc fBodyBodyAccJerkMag fBodyBodyGyroMag angle(Z,gravityMean) fBodyAccJerk angle(tBodyAccMean,gravity) tGravityAccMag tBodyGyro tBodyGyroJerkMag angle(X,gravityMean) angle(tBodyGyroJerkMean,gravityMean) tBodyGyroJerk fBodyBodyGyroJerkMag
In [64]:
for stat in static_methods:
print(stat + "\n")
skewness() meanFreq() maxInds max() correlation() arCoeff() arCoeff()3 bandsEnergy() arCoeff()4 min() mad() sma() iqr() std() entropy() energy() arCoeff()2 arCoeff()1 mean() kurtosis()
Funkcje do estymacji sygnałów¶
- max()
- mad() - mediana odchylenia bezwzględnego
- min()
- kurtosis() - jedna z miar kształtu rozkładu częstotliwości sygnału
- bandsEnergy() - energia przedziału częstotliwości w przedziałach FFT każdego okna
- mean()
- meanFreq()
- arCoeff() - współczynnik autoregresji
- entropy() - entropia sygnału, średnia ilość informacji przypadająca na pojedynczą wiadomość ze źródła
- iqr() - $Q_3 - Q_1 $
- sma() - obszar wielkości sygnału
- std()
- maxInds - indeks składowej częstotliwości o największej wartości
- skewness()
- energy()
- correlation()
Patrzymy na rozkłady dla interesujących nas statystyk
In [65]:
stat = 'mean'
stat_cols = [col for col in data.columns if stat in col]
In [66]:
data[stat_cols].hist(figsize = (50,50))
Out[66]:
array([[<Axes: title={'center': 'tBodyAcc-mean()-X'}>,
<Axes: title={'center': 'tBodyAcc-mean()-Y'}>,
<Axes: title={'center': 'tBodyAcc-mean()-Z'}>,
<Axes: title={'center': 'tGravityAcc-mean()-X'}>,
<Axes: title={'center': 'tGravityAcc-mean()-Y'}>,
<Axes: title={'center': 'tGravityAcc-mean()-Z'}>,
<Axes: title={'center': 'tBodyAccJerk-mean()-X'}>],
[<Axes: title={'center': 'tBodyAccJerk-mean()-Y'}>,
<Axes: title={'center': 'tBodyAccJerk-mean()-Z'}>,
<Axes: title={'center': 'tBodyGyro-mean()-X'}>,
<Axes: title={'center': 'tBodyGyro-mean()-Y'}>,
<Axes: title={'center': 'tBodyGyro-mean()-Z'}>,
<Axes: title={'center': 'tBodyGyroJerk-mean()-X'}>,
<Axes: title={'center': 'tBodyGyroJerk-mean()-Y'}>],
[<Axes: title={'center': 'tBodyGyroJerk-mean()-Z'}>,
<Axes: title={'center': 'tBodyAccMag-mean()'}>,
<Axes: title={'center': 'tGravityAccMag-mean()'}>,
<Axes: title={'center': 'tBodyAccJerkMag-mean()'}>,
<Axes: title={'center': 'tBodyGyroMag-mean()'}>,
<Axes: title={'center': 'tBodyGyroJerkMag-mean()'}>,
<Axes: title={'center': 'fBodyAcc-mean()-X'}>],
[<Axes: title={'center': 'fBodyAcc-mean()-Y'}>,
<Axes: title={'center': 'fBodyAcc-mean()-Z'}>,
<Axes: title={'center': 'fBodyAcc-meanFreq()-X'}>,
<Axes: title={'center': 'fBodyAcc-meanFreq()-Y'}>,
<Axes: title={'center': 'fBodyAcc-meanFreq()-Z'}>,
<Axes: title={'center': 'fBodyAccJerk-mean()-X'}>,
<Axes: title={'center': 'fBodyAccJerk-mean()-Y'}>],
[<Axes: title={'center': 'fBodyAccJerk-mean()-Z'}>,
<Axes: title={'center': 'fBodyAccJerk-meanFreq()-X'}>,
<Axes: title={'center': 'fBodyAccJerk-meanFreq()-Y'}>,
<Axes: title={'center': 'fBodyAccJerk-meanFreq()-Z'}>,
<Axes: title={'center': 'fBodyGyro-mean()-X'}>,
<Axes: title={'center': 'fBodyGyro-mean()-Y'}>,
<Axes: title={'center': 'fBodyGyro-mean()-Z'}>],
[<Axes: title={'center': 'fBodyGyro-meanFreq()-X'}>,
<Axes: title={'center': 'fBodyGyro-meanFreq()-Y'}>,
<Axes: title={'center': 'fBodyGyro-meanFreq()-Z'}>,
<Axes: title={'center': 'fBodyAccMag-mean()'}>,
<Axes: title={'center': 'fBodyAccMag-meanFreq()'}>,
<Axes: title={'center': 'fBodyBodyAccJerkMag-mean()'}>,
<Axes: title={'center': 'fBodyBodyAccJerkMag-meanFreq()'}>],
[<Axes: title={'center': 'fBodyBodyGyroMag-mean()'}>,
<Axes: title={'center': 'fBodyBodyGyroMag-meanFreq()'}>,
<Axes: title={'center': 'fBodyBodyGyroJerkMag-mean()'}>,
<Axes: title={'center': 'fBodyBodyGyroJerkMag-meanFreq()'}>,
<Axes: >, <Axes: >, <Axes: >]], dtype=object)
widać duzo róznych rozkładów.
3. macierz korelacji¶
3.1 mean¶
In [67]:
plt.figure(figsize=(40,40))
sns.heatmap(data[stat_cols].corr(),annot=True, cmap='coolwarm')
plt.title('Correlation matrix for "mean"')
plt.show()
bardzo duże korelacje niektórych kolumn, czy je usuwać? (z doświadczenia usuwanie to nie jest najlepsza opcja)
3.2 entropy¶
In [68]:
stat = 'entropy'
stat_cols1 = [col for col in data.columns if stat in col]
In [69]:
plt.figure(figsize=(40,40))
sns.heatmap(data[stat_cols1].corr(),annot=True, cmap='coolwarm')
plt.title('Correlation matrix for "entropy"')
plt.show()
bardzo duze korelacje entropii. Moze nie jest ona potrzebna/znacząca?
4. Pairplot¶
4.1 mean¶
In [70]:
sns.pairplot(data[stat_cols[:10]])
Out[70]:
<seaborn.axisgrid.PairGrid at 0x30bf7a490>
widać czasami jakiś podział
In [71]:
#save plot
plt.savefig('../EDA/pairplot.png')
<Figure size 640x480 with 0 Axes>
4.2 entropy¶
In [72]:
sns.pairplot(data[stat_cols1[:10]])
Out[72]:
<seaborn.axisgrid.PairGrid at 0x321f83e90>
widać podział mimo wysokich korelacji
5. boxplot¶
5.1 mean¶
In [73]:
fig, axes = plt.subplots(10, 5, figsize=(40, 50))
axes = axes.flatten()
for i, col in enumerate(stat_cols):
sns.boxplot(data, x= data[col], ax= axes[i])
plt.tight_layout()
plt.show()
5.2 entropy¶
In [74]:
fig, axes = plt.subplots(7, 5, figsize=(40, 50))
axes = axes.flatten()
for i, col in enumerate(stat_cols1):
sns.boxplot(data, x= data[col], ax= axes[i])
plt.tight_layout()
plt.show()
6. TSNE¶
In [75]:
#tsne
from sklearn.manifold import TSNE
import matplotlib.pyplot as plt
import seaborn as sns
tsne = TSNE(n_components=2, random_state=42)
tsne_results = tsne.fit_transform(data)
#plot results
plt.figure(figsize=(10,5))
sns.scatterplot(x=tsne_results[:,0], y=tsne_results[:,1])
Out[75]:
<Axes: >
widać cos na kształt klastra
tam po prawej troche 'ni pies ni wydra, cos na ksztalt świdra'